The present invention relates generally to automatic performance devices, and more particularly to automatic performance devices which, during a first automatic performance, allows a second automatic performance to be added in response to operation of a switch or the like.
The present invention further relates to automatic performance devices which is capable of reproducing a phrase by reading out prestored phrase performance data.
Automatic performance devices are known in the art, where an operator enters performance data corresponding to individual musical notes by means of real-time or step recording, and the entered performance data are then reproduced to execute an automatic performance. In addition to such devices, there are known another type of automatic performance devices in which automatic performance data made of short phrases (hereinafter referred to as phrase data) are previously assigned to a plurality of switches, in order to achieve varied automatic performances by reading out the phrase data in response to operation of any of the switches and thereby adding a phrase performance to a normal automatic performance.
For instance, Japanese Patent Laid-open Publication No. HEI 4-356100 that corresponds to U.S. Pat. No.5,281,756 discloses an automatic performance device which prestores automatic performance data made of predetermined phrases in correspondence to operators or operating members such as switches, and in which the prestored automatic performance data is read out for reproduction instantly in response to operation of any of the operating members. In such an automatic performance device, it is possible to, during a manual performance on a keyboard or the like or an automatic performance (which is different from the above-mentioned automatic phrase performance and will hereinafter be referred to as a first automatic performance), add an automatic performance of predetermined phrases (hereinafter referred to as a second automatic performance) to the first automatic performance.
However, with such prior art automatic performance devices, the beat timings of the second automatic performance often do not properly coincide with those of the first automatic performance. Namely, because as mentioned earlier, the second automatic performance is arranged to get started immediately upon operation of the operating member, the beat timings of the first and second automatic performance will considerably differ from each other unless the user or player operates the operating member to the beat timings of the first performance. But, it is not at all easy for the player to make his or her operation timings of the operating member exactly coincident with the first performance beat timings.
Further with the prior art automatic performance devices, a phrase performance is achieved only by the player's real-time operation of the operating member, so that there arises a problem that a same automatic performance with a phrase performance added thereto can be executed only once but can never be repeated.
U.S. Pat. No. 5.155,286 also discloses that automatic performance is executed by repetitively reading out prestored motif performance data while a predetermined tone-generation instructing switch is operated. However, the U.S. Patent fails to disclose a technique of assigning a selected one of plural performance data to an operating member and recording operation data indicative of operation of the operating member, nor does it disclose a technique of starting a motif performance at a predetermined timing (beat) after operation of the operating member.